Flexible palm rehabilitation training system and method based on multi-mode sensing and self power supply

Abstract

The invention discloses a flexible palm rehabilitation training system and method based on multi-mode sensing and self power supply. The system comprises a pneumatic system, a palm heat power supply system, an energy storage unit and a main control unit. The pneumatic system comprises a glove body, an air pump and a pipeline, the air pump communicates with the pipeline, the pipeline is fixed to the glove body, the air pump and the main control unit are both connected with the energy storage unit, the air pump is in communication connection with the main control unit, and the energy storage unit is connected with the palm heat power supply system; the palm thermal power supply system can convert skin temperature into electric energy based on the thermoelectric effect to supply power to the energy storage unit. A self-sustainable energy supply system, a high-precision multi-dimensional sensing network and a flexible pneumatic driving interaction module are integrated to assist hand rehabilitation training.

Description

Technical Field

[0001] The present invention relates to the technical field of medical rehabilitation equipment, and in particular to a flexible palm rehabilitation training system and method based on multimodal sensing and self-power supply. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Patients with hand dysfunction need to use hand rehabilitation equipment for hand rehabilitation training. Currently, there are several types of exoskeleton gloves: (1) A flexible exoskeleton glove system for hand rehabilitation training, comprising a finger data acquisition glove, a decision-making drive module, and a flexible exoskeleton glove. The data glove comprises a first body, a flexible sensor, and a wireless transmission module; the decision-making drive comprises a decision-making and drive modules, the former comprising a microcontroller and a wireless receiving module, and the latter comprising a support frame, an air pump, and an electric proportional valve; the flexible exoskeleton glove comprises a second body and a flexible bending actuator (comprising Kevlar fiber line, silicone tube, glass fiber cloth, and pneumatic seal).

[0004] (2) A variable stiffness rigid-flexible coupling hand rehabilitation exoskeleton based on wire drive. The exoskeleton contains a flexible textile glove, each fingertip has a phalange, and is connected to a traction rope and a pull rope to drive bending and straightening. The traction rope is located on both sides of the fingertip near the palm and is connected to the servo assembly to provide bending force; the pull rope is an elastic rope located in the middle of the fingertip near the back of the hand to provide straightening and restoring force.

[0005] (3) An intelligent exoskeleton glove and a method for using the same, comprising a flexible glove and a drive mechanism, the glove having a palm portion and at least three finger portions, each finger portion being equipped with a pressure sensor, a bending sensor, and a drive cord. The drive mechanism is electrically connected to the sensor and to the cord.

[0006] It can be seen that existing finger rehabilitation equipment mostly adopts rigid mechanical structures, which are prone to secondary damage and have poor wearing comfort; traditional power supply methods rely on external power supplies or batteries, and have problems such as insufficient battery life, large size and weight, and inconvenience in carrying; single modality sensors cannot meet the multi-dimensional monitoring needs of complex rehabilitation training, and temperature and humidity monitoring has the defect of insufficient spatial resolution; existing pneumatic control systems generally use rigid pneumatic components, which lack flexibility and adaptive adjustment capabilities. Summary of the Invention

[0007] In order to solve the above problems, the present invention proposes a flexible palm rehabilitation training system and method based on multimodal sensing and self-powered power supply, which assists hand rehabilitation training by integrating a self-sustainable energy supply system, a high-precision multidimensional sensing network and a flexible pneumatic drive interaction module.

[0008] To achieve the above object, the present invention adopts the following technical solutions: First, a flexible palm rehabilitation training system based on multimodal sensing and self-power supply is proposed, including a pneumatic system, a palm thermal power supply system, an energy storage unit, and a main control unit. The pneumatic system includes a glove body, an air pump, and a pipeline. The air pump is connected to the pipeline, which is fixed to the glove body. The air pump and the main control unit are both connected to the energy storage unit. The air pump and the main control unit are in communication connection, and the energy storage unit is connected to the palm heat power supply system. The palm thermal power supply system can convert skin temperature into electrical energy based on the thermoelectric effect to power the energy storage unit.

[0009] Furthermore, it also includes a multimodal sensing system, which is connected to the palm thermal power supply system and the main control unit; The multimodal sensing system is used to obtain the temperature and humidity of each area of ​​the glove body; The main control unit is used to issue an alarm message when the temperature or humidity exceeds the set threshold.

[0010] Furthermore, the main control unit is used to issue a third-level alarm message when the humidity is greater than a first set humidity value, or the temperature is greater than a first set temperature threshold, and the duration is greater than a set time; when the temperature is greater than a second set temperature threshold, it is used to issue a second-level alarm message.

[0011] Furthermore, the main control unit is also used to determine that when the humidity of a certain part is greater than the first set humidity value, based on the temperature and humidity of the part, the reason is the increased humidity. Only when the reason for the increased humidity is physiological sweating will a third-level alarm message be issued.

[0012] Furthermore, the thermoelectric power generation layer includes a flexible substrate and a TEG unit array provided on the flexible substrate; the TEG unit array in the thermoelectric power generation layer includes three connection layers; wherein the first connection layer includes two connected TEG units, and a switch S is further provided on the connection circuit of the two TEG units. P1A and S S1A , by controlling the switch S P1A and S S1A The second connection layer includes two connected first connection layers, and a switch S is set on the two connected first connection layers. P2A and S S2A , by controlling the switch S P2A and S S2A The two first connection layers are connected in series or in parallel by opening and closing; The third connection layer includes three connected second connection layers, on which switches S are set. P3A 、S S3A and S S3B , by controlling the switch SP3A 、S S3A and S S3B The three second connection layers are connected in series or in parallel by switching them on and off.

[0013] Furthermore, it also includes a wristband, on which the air pump, energy storage unit and main control unit are all installed. A display screen is also provided on the wristband, and the display screen is connected to the main control unit.

[0014] Furthermore, the main control unit is connected to the mobile client.

[0015] Furthermore, the palm thermal power supply system includes a contact thermal conductive layer, a thermoelectric generator array layer, a heat dissipation layer and a packaging layer stacked in sequence; wherein the contact thermal conductive layer is used to contact the skin.

[0016] Furthermore, the pipeline includes five finger pipelines; the air pump includes five finger air pumps and a main air pump; the five finger pipelines are connected to the five finger air pumps in a one-to-one correspondence; and the five finger pipelines are also connected to the main air pump.

[0017] Secondly, a flexible palm rehabilitation training method based on multimodal sensing and self-powered power supply is proposed, including: The palm heat power supply system is based on the thermoelectric effect, converting skin temperature into electrical energy to power the energy storage unit; Power the air pump and main control unit through the energy storage unit; The main control unit controls the air pump to inflate the pipes, thereby controlling the movement of the glove body.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The present invention proposes a flexible palm rehabilitation training system and method based on multimodal sensing and self-powering. The system is equipped with a palm thermal power supply system, which can convert skin temperature into electrical energy to power the energy storage unit, and then power the pneumatic system and main control unit through the energy storage unit without relying on an external power supply, thereby improving the system's endurance. By providing an air pump and pipelines, the air pump can realize the inflation and deflation of the pipelines, thereby controlling the movement of each finger in the glove body, which can better fit the human hand and greatly reduce the risk of skin compression compared to rigid mechanical devices. This solves the technical problems of current palm rehabilitation equipment such as reliance on external power supply, the existing rigid structure that is prone to secondary damage, and the lack of flexibility of pneumatic control.

[0019] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings in the specification, which constitute a part of this application, are used to provide further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute improper limitations on this application.

[0021] Figure 1 This is a schematic diagram of the overall structure of the flexible palm rehabilitation training system proposed in the present invention; Figure 2 This is a schematic diagram of the Seebeck effect principle of the present invention; Figure 3 This is a circuit diagram of the thermoelectric power generation layer of the present invention; Figure 4 This is a schematic diagram of the palm heat power supply circuit of the present invention; Figure 5 is a diagram of the arrangement of the bending sensor of the present invention; Figure 6 It is the layout diagram of the temperature and humidity sensor of the present invention; Figure 7 This is a layered structure diagram of the temperature and humidity sensor of the present invention; Figure 8 This is an assembly diagram of the integrated wristband of the present invention; Figure 9 It is a software end interface diagram of the present invention; Figure 10 It is the overall workflow diagram of the present invention.

[0022] Ⅰ-1, packaging layer, Ⅰ-2, heat dissipation layer, Ⅰ-3, thermoelectric power generation layer, Ⅰ-4, contact thermal conductive layer; Ⅱ-1. Triboelectric sensor, Ⅱ-2. Capacitive bending sensor, Ⅱ-3. Auxiliary IMU sensor, Ⅱ-4. Flexible precision temperature and humidity sensor, Ⅱ-4-1. Circular flexible packaging layer, Ⅱ-4-2. Circular flexible thermal material upper electrode, Ⅱ-4-3. Circular flexible thermal material, Ⅱ-4-4. Circular flexible thermal material lower electrode, Ⅱ-4-5. Circular flexible humidity-sensitive material upper electrode, Ⅱ-4-6. Circular flexible humidity-sensitive material, Ⅱ-4-7. Circular flexible humidity-sensitive material lower electrode, Ⅱ-4-8. Flexible base layer, Ⅱ-5. Thin film NTC temperature and humidity sensor Sensor, II-5-1, rectangular packaging layer, II-5-2, rectangular upper electrode, II-5-3, rectangular temperature-sensitive functional layer, II-5-4, rectangular humidity-sensitive functional layer, II-5-5, rectangular lower electrode, II-5-6, rectangular substrate, II-6, environmental temperature and humidity sensor, II-6-1, strip packaging layer, II-6-2, strip thermal conductive layer, II-6-3, strip thermistor, II-6-4, strip humidity-sensitive material upper electrode, II-6-5, strip humidity-sensitive material, II-6-6, strip humidity-sensitive material lower electrode, II-6-7, strip base layer, II-8, diversion gap; Ⅲ-1, pipeline, Ⅲ-2, connection, Ⅲ-3, glove body; IV. Integrated bracelet, IV-1. Display screen, IV-2. Main control unit, IV-3. Battery, IV-4. Wrist strap, IV-5. Air pump. DETAILED DESCRIPTION

[0023] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0024] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present application belongs.

[0025] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0026] In the present invention, terms such as "fixed connection," "connected," and "connection" should be interpreted broadly to mean a fixed connection, an integral connection, or a detachable connection; a direct connection or an indirect connection through an intermediary. Relevant researchers or technicians in this field may determine the specific meanings of these terms in the present invention based on specific circumstances, and they should not be construed as limitations of the present invention.

[0027] Example 1 In this embodiment, a flexible palm rehabilitation training system based on multimodal sensing and self-power supply is disclosed, including a pneumatic system, a palm thermal power supply system, an energy storage unit and a main control unit; The pneumatic system includes the glove body III-3, an air pump IV-5, and a pipeline III-1. The air pump IV-5 is connected to the pipeline III-1, which is fixed to the glove body III-3. The air pump IV-5 and the main control unit are both connected to the energy storage unit. The air pump IV-5 is in communication with the main control unit, and the energy storage unit is connected to the palm heat power supply system. The palm thermal power supply system can convert skin temperature into electrical energy based on the thermoelectric effect to power the energy storage unit.

[0028] The system proposed in this embodiment further includes a multimodal sensing system, which is connected to both the palm thermal power supply system and the main control unit; The multimodal sensing system is used to obtain the temperature and humidity of each area of ​​the glove body; The system proposed in this embodiment also includes a wristband, and the air pump, energy storage unit and main control unit are all installed on the wristband. A display screen is also provided on the wristband, and the display screen is connected to the main control unit.

[0029] The main control unit is used to issue an alarm message when the temperature or humidity exceeds the set threshold.

[0030] Among them, the multimodal sensing system is also used to obtain the pressure and movement status of the finger; The main control unit is used to control the air pump according to the pressure and movement status of the finger.

[0031] The palm thermal power supply system includes a contact thermal conductive layer, a thermoelectric generator array layer, a heat dissipation layer and a packaging layer stacked in sequence; wherein the contact thermal conductive layer is used to contact the skin.

[0032] Combine Figures 1-10 , a detailed description is given of the flexible palm rehabilitation training system based on multimodal sensing and self-powered power disclosed in the embodiment of the present application.

[0033] The flexible palm rehabilitation training system based on multimodal sensing and self-power supply disclosed in the embodiment of the present application includes: a pneumatic system, a palm thermal power supply system, a multimodal sensing system and an integrated wristband IV.

[0034] Among them, the palm heat power supply system is as follows Figure 1 As shown, the outermost layer is the biocompatible encapsulation layer I-1, and the innermost layer is the contact thermal conductive layer I-4. Contact thermal conductive layer I-4 adheres closely to the skin, sensing the palm's skin temperature and ensuring power supply stability. Thermoelectric generation layer I-3 is located between contact thermal conductive layer I-4 and encapsulation layer I-1. A heat dissipation layer I-2 is also located between these two layers. Thermoelectric generation layer I-3 utilizes the Seebeck effect to convert heat energy transferred from contact thermal conductive layer I-4 into electrical energy, which is then output. Thermoelectric generation layer I-3 is located between contact thermal conductive layer I-4 and heat dissipation layer I-2 to prevent overheating inside the glove.

[0035] Preferably, the thermoelectric power generation layer I-3 includes a flexible substrate and a TEG (thermoelectric generator) unit array arranged on the flexible substrate, each TEG unit includes an N-type semiconductor and a P-type semiconductor. When there is a temperature difference in the closed loop composed of the N-type semiconductor and the P-type semiconductor, a potential difference (voltage) will be generated at both ends.

[0036] like Figure 3 As shown, the TEG unit array in the thermoelectric power generation layer I-3 includes three connection layers; wherein the first connection layer includes two connected TEG units, and a switch S is also set on the connection circuit of the two TEG units. P1A and S S1A , by controlling the switch S P1A and SS1A By switching on and off, two TEG units can be connected in series or in parallel.

[0037] The second connection layer includes two connected first connection layers, and switches S are set on the two connected first connection layers. P2A and S S2A , by controlling the switch S P2A and S S2A By switching on and off, the two first connection layers are connected in series or in parallel.

[0038] The third connection layer includes three connected second connection layers, on which switches S are set. P3A 、S S3A and S S3B , by controlling the switch S P3A 、S S3A and S S3B The three second connection layers are connected in series or in parallel by switching on and off. P3A 、S S3A Disconnect, S S3B When closed, the three second connection layers are connected in parallel; when S P3A 、S S3A Disconnect, S S3B When all are disconnected, the three second connection layers are connected in series.

[0039] Through the coordinated control of the three connection layers, it is suitable for different application scenarios. Specifically, when all 12 TEG units are connected in series, the thermoelectric power generation layer I-3 can output high voltage, V out =12×V TEG , suitable for low-power devices; when two groups of TEG cells in the 12 TEG cells are connected in series and six TEG cells in each group are connected in parallel, the voltage and current can be balanced and adapted to medium loads; when three groups of TEG cells in the 12 TEG cells are connected in series and four TEG cells in each group are connected in parallel, the power output can be optimized and the internal resistance voltage drop of the thermocouple can be reduced; when four groups of TEG cells in the 12 TEG cells are connected in series and three TEG cells in each group are connected in parallel, the dynamic response speed can be improved and it is suitable for load mutation scenarios; when six groups of TEG cells in the 12 TEG cells are connected in series and two TEG cells in each group are connected in parallel, the current capacity can be increased and multiple devices can be supported in parallel; when all 12 TEG cells are connected in parallel, the output current can be maximized to achieve fast charging or energy recovery.

[0040] Figure 2The N in the figure represents an N-type semiconductor, and the P represents a P-type semiconductor. The ends of the N-type semiconductor and the P-type semiconductor are both connected to the contact thermal conductive layer I-4, and the heat energy transferred from the contact thermal conductive layer I-4 is used as the heat source. One end of the heat dissipation layer I-2 is used as the cold end. The P-type semiconductor is doped with acceptor impurities (such as boron and aluminum) to form positively charged holes as the main carriers. After absorbing heat at the hot end, the holes diffuse to the cold end, forming the positive direction component of the current; the N-type semiconductor is doped with donor impurities (such as phosphorus and arsenic) to form negatively charged free electrons as the main carriers. After releasing heat at the hot end, the electrons diffuse to the cold end, forming the negative direction component of the current.

[0041] The principle is as follows Figure 2 This diagram illustrates the core principle of the thermoelectric effect (Seebeck effect): a heat source creates a temperature gradient, causing electrons (negatively charged) in the N-type semiconductor (left) and holes (positively charged) in the P-type semiconductor (right) to diffuse toward the cold end. This heat builds up charge at the cold end (negative at the N-end and positive at the P-end), generating a built-in electric field that points from positive to negative. When the external circuit is closed, the electric field drives electrons through the wire from the N-type cold end to the P-type cold end, generating thermoelectric power and achieving direct conversion of thermal energy into electrical energy.

[0042]

[0043] S: Seebeck coefficient (material property, unit: μV / K) ΔT: Temperature difference between hot end and cold end (unit: K) Thermoelectric conversion efficiency formula:

[0044] ZT m : dimensionless figure of merit of material; η: thermoelectric conversion efficiency (unitless, usually expressed as a percentage); T h : hot end temperature (unit: K, Kelvin); T e : Cold junction temperature (unit: K, Kelvin).

[0045] The circuit principle diagram of the thermoelectric power generation layer proposed in this embodiment is as follows Figure 4As shown in the figure, this circuit includes a power management IC (U1). Pins 1-4 of U1 are configuration and control pins; pins 5-6 are power input connections; pin 7 is the enable or control output; pins 8-9 are reference voltage or internal regulator connections; pin 10, labeled "OK_HYST," sets the hysteresis threshold for the power OK signal; pin 11 is the charge status output; and pins 12-13 are power outputs. AD_VBAT is an analog voltage output for monitoring battery voltage; and VBAT_OV is the battery overvoltage protection signal.

[0046] Pin 16 is the LBST terminal, connecting the TEG cell array on the thermoelectric generation layer to Zener diode D1. The TEG cells generate weak DC power using ambient temperature differences. This power is protected by Zener diode D1 and then fed into U1. Zener diode D1 provides reverse power protection for the TEG cell array, preventing damage to the circuit due to reverse power connection. Inductor L1 is connected in series between Zener diode D1 and pin 16 of U1, filtering and storing energy. One end of capacitor C1 is connected between Zener diode D1 and inductor L1, with the other end of capacitor C1 grounded. Capacitor C1 filters high-frequency noise. A connection port P1 is also connected between Zener diode D1 and capacitor C1, which connects to a connector, serving as the power output interface. This circuit serves as an input protection circuit.

[0047] One end of resistor R2, pins 2 and 3 of U1 are connected between capacitor C1 and inductor L1; the other end of resistor R2 is connected to resistor R3 and pin 3 of U1, and resistor R3 is grounded; pin 4 of U1 is connected to capacitor C2; capacitor C2 is grounded. Pin 6 of U1 is connected to one end of resistor R9 and one end of resistor R4; the other end of resistor R9 is grounded; the other end of resistor R4 is connected to pin 7 of U1, and pin 7 of U1 is also connected to one end of resistor R5 and one end of resistor R6; the other end of resistor R5 is connected to one end of resistor R10, and resistor R10 is grounded; the other end of resistor R6 is connected to resistor R11, and resistor R11 is connected to resistor R8, and resistor R8 is grounded; pin 8 of U1 is connected to the circuit connecting resistor R6 and pin 7 of U1 and to the circuit connecting resistor R5 and resistor R10; pin 9 of U1 is connected between resistor R6 and resistor R11; pin 10 of U1 is connected between resistor R11 and resistor R8; pin 14 of U1 is grounded via capacitor C6 and capacitor C7; capacitor C6 and capacitor C7 are connected in parallel; pin 5 of U1 is grounded; pin 15 of U1 is grounded via capacitor C3, capacitor C4 and capacitor C5; pins 1, 12, 13 and 17 of U1 are all grounded. Capacitors C3, C4, and C5 ensure output voltage stability. Resistors R4, R5, ..., R9 provide voltage division or current limiting. Resistors R10 and R11 are used for maximum power point tracking (MPPT) control, and capacitors C6 and C7 provide additional filtering to ensure power quality.

[0048] The voltage-dividing circuit formed by resistors R1 and R2, called a voltage-sensing network, monitors the output voltage of the thermoelectric generator (TEG). This network is connected to one of U1's analog inputs (pins 5 or 6), allowing U1 to obtain real-time voltage information from the TEG.

[0049] Resistors R8 and R9 form a current sensing network that monitors the output current of the thermoelectric layer. By measuring the voltage drop across this resistor, U1 can calculate the output current.

[0050] U1 uses the obtained output voltage and output current to dynamically adjust the output voltage and output current through the internal MPPT algorithm, so that the TEG always operates near the maximum power point.

[0051] This design enables the system to adapt to changes in the output characteristics of the TEG due to temperature differences, ensuring that maximum energy can be obtained from the TEG under different environmental conditions. At the same time, the supercapacitors (C1 to C7) store excess energy, reducing the burden on the main battery and extending its life. This circuit is designed specifically for thermoelectric energy harvesting systems, collecting the electrical energy generated by the thermoelectric generation layer. It uses input protection circuitry to ensure safe and reliable energy input. U1 implements battery charging control and power path management. It provides a stable output voltage through a filter circuit, monitors battery status, and provides system status feedback through signals such as VBAT_OK. This design adapts to changes in the TEG's output characteristics due to temperature differences, ensuring maximum energy extraction from the TEG under varying environmental conditions. Meanwhile, supercapacitors (C1 to C7) store excess energy, reducing the burden on the main battery and extending its life.

[0052] The TEG cell array on the thermoelectric generation layer is connected to Zener diode D1. The TEG cells utilize ambient temperature differences to generate weak DC power, which is protected by Zener diode D1 before being fed into U1. Next, cold start and boost are implemented: When the voltage input to U1 exceeds the set voltage, the boost converter initiates a cold start, raising the input voltage to the threshold voltage and activating the boost circuit, ultimately outputting a stable voltage. MPPT optimization is also performed: a resistor network dynamically adjusts the input impedance to track the TEG maximum power point. Next, energy storage management is implemented: to ensure the daily use of the rehabilitation glove, supercapacitors (capacitors C1-C7) store energy, charging to the rated voltage cutoff. When the voltage drops below a certain level, the device automatically shuts down and provides power to the system when the voltage is normal. Finally, power output is implemented: U1 converts the supercapacitor voltage to 3.3V (the voltage required by the sensor). The supercapacitor smooths voltage fluctuations in the TEG output, such as power jitter caused by changes in human body temperature, preventing the battery from experiencing frequent high-current charge and discharge, thereby improving system efficiency and battery life.

[0053] By dynamically adjusting the input impedance of the TEG array so that it always matches the optimal load under the current heat source conditions, the thermal energy conversion efficiency is maximized under a wide input voltage range (0.15V~10.8V) and complex environmental changes, and a stable and efficient energy supply is provided for multiple output loads (1.2V / 2.7V / 3.6V) and energy recovery mechanisms.

[0054] Supercapacitors can be used as buffer storage for energy recovery mechanisms: When the TEG output power temporarily exceeds the load demand, the excess energy is stored in the supercapacitor through U1; when the power is insufficient, it is released back to the system. This reduces the charge and discharge pressure of the battery and extends its life.

[0055] The battery is the main energy source of the system: it stores the accumulated energy collected by TEG over a long period of time (such as several hours) and provides continuous power supply for low-power sensors, MCU, etc.

[0056] Supercapacitors can smooth out voltage fluctuations output by TEGs, such as power jitter caused by changes in human body temperature, and prevent the battery from frequently experiencing high current charging and discharging, thereby improving system efficiency and battery life.

[0057] Example of a complete workflow for thermoelectric generation layer I-3: 1. Startup: TEG thermoelectric generator → Boost converter cold start → Boost to rated voltage → Supercapacitor charging.

[0058] 2. Power supply: Supercapacitor voltage ≥ preset voltage V → boost to 3.3V → data acquisition 3. Sleep: Data upload successful → Return to sleep mode (RTC timing).

[0059] like Figure 5 As shown, the flexible palm rehabilitation training system based on multimodal sensing and self-power supply proposed in this embodiment also includes a multimodal sensing system, which is connected to the palm thermal power supply system and the main control unit; The multimodal sensing system is used to obtain the temperature and humidity of each area of ​​the glove body; The main control unit is used to issue an alarm message when the temperature or humidity exceeds the set threshold.

[0060] In addition, the multimodal sensing system is also used to obtain the pressure and movement status of the finger; The main control unit is used to control the air pump according to the pressure and movement status of the finger.

[0061] Specifically, the multimodal sensing system includes capacitive bending sensor II-2, triboelectric sensor II-1, auxiliary IMU sensor II-3, flexible precision temperature and humidity sensor II-4, thin film NTC temperature and humidity sensor II-5, ambient temperature and humidity sensor II-6, Among them, the capacitive bending sensor II-2 and the triboelectric sensor II-1 are installed on the fingers of the glove body. In order to monitor the bending angle and movement state of the fingers, the core sensor selects the capacitive bending sensor II-2, which is used to monitor the bending angle of the fingertips; this sensor supports parallel measurement of multiple fingers; the triboelectric sensor II-1 is used for dynamic continuous monitoring, and can eliminate signal attenuation through the charge retention circuit, making it suitable for long-term motion capture. Its self-powered characteristics can reduce the power consumption of the gloves and extend the battery life to 72 hours; the auxiliary IMU sensor II-3 is installed on the back of the hand of the glove body to obtain the angular velocity of the back of the hand; posture calibration and complex gesture recognition are performed through the angular velocity of the motion.

[0062] In terms of data processing, an embedded intelligent AI chip uses a Kalman filter algorithm to fuse the bend sensor (angle data) with the auxiliary IMU sensor (angular velocity data), improving measurement accuracy in dynamic scenarios. To further ensure measurement accuracy, a dynamic calibration mechanism is introduced, performing a zero-position calibration every five minutes, determining the reference angle using the fully extended finger position.

[0063] Table 1 Noise covariance matrix

[0064] Table 1 quantifies the statistical characteristics of system noise and measurement noise, providing key parameters for the Kalman filter algorithm to ensure the accuracy and stability of state estimation (such as bending angle and angular velocity) in complex rehabilitation training scenarios.

[0065] By properly setting Q and R, the filter can distinguish between sensor noise (high R) and model uncertainty (high Q), avoid over-reliance on unreliable data, and suppress noise interference.

[0066] Improve estimation robustness. For example, the IMU's integration drift (low-frequency noise) is compensated by the low-frequency components of the Q matrix, while the bending sensor's random noise (high-frequency noise) is suppressed by the R matrix.

[0067] Optimize computational efficiency: The diagonal matrix structure simplifies covariance operations, reduces computational complexity, and meets real-time requirements.

[0068] The first step is to build the system model, the state vector:

[0069] is the state vector at time point k; θ k is the real-time bending angle; is the angular velocity (provided by IMU); State transition equation:

[0070] Δt: sampling period; w k : process noise (zero-mean Gaussian noise, covariance matrix Q); Measurement equation:

[0071] z k : bending sensor measurement value; v k : measurement noise (covariance matrix R); Dynamic scenes use a variable step size strategy:

[0072] Motion compensation mechanisms include angular velocity compensation: automatically reducing the Q(2,2) weight (to 0.0001) when the IMU detects high-frequency vibration (>20Hz), and acceleration assistance: using IMU accelerometer data to detect stationary state (freezing angular velocity updates when |α|<0.1g) The extended Kalman filter (EKF) is introduced to handle the nonlinearity of the model when the angle is large (>90°):

[0073] Where h is the measurement function; θ is the state variable (such as attitude angle deviation); is the angular velocity; Δt is the time update interval, the time difference between two consecutive observations.

[0074] In terms of temperature monitoring, Figure 6 As shown, to detect thermal damage caused by temperature, temperature and humidity sensors are distributed throughout the glove body, with flexible precision temperature and humidity sensors II-4 placed on the fingers. The temperature and humidity sensors are arranged in a dense and sparse design principle: a high-density, high-precision core area and a sparse, low-consumption peripheral area. The core area extends to the palm, and the peripheral area refers to the area outside the core area.

[0075] Fingertips / palms: 5-8 mm spacing, high-precision sensors + real-time warnings.

[0076] Joints / back of hand: 10-20 mm spacing, taking into account both mechanical flexibility and cost.

[0077] Algorithm compensation: fill monitoring blind spots in sparse areas through data fusion High-density placement on fingertips: 2-3 flexible precision temperature sensors II-4 (with spacing of approximately 8-10 mm) are placed on the fingertips of the index finger, middle finger, ring finger, and little finger. Because the fingertips are most likely to come into contact with high-temperature objects, local temperature gradient changes need to be captured.

[0078] Thin-film NTC sensors II-5 (15-20 mm spacing) are distributed in the palm area (such as the center of the palm and the thenar area) to monitor the overall temperature distribution through heat conduction, avoiding overlap with the pressure sensor and affecting measurement accuracy.

[0079] Wrist auxiliary monitoring point: An ambient temperature sensor II-6 is installed on the wrist crease of the glove body to obtain the ambient temperature. By monitoring the ambient temperature, it can distinguish between ambient temperature rise and direct contact with the heat source. This solution can reduce the false alarm rate by 40%.

[0080] In terms of humidity sensors, such as Figure 6 As shown in the figure, a core monitoring area layout is adopted to establish a three-dimensional gradient monitoring network, and a three-level monitoring system of "fingertip-palm-wrist" is constructed inside the glove: 2-3 flexible precision temperature and humidity sensors II-4 are arranged in the fingertip area of ​​each finger. Since sweat secretion is most active at the fingertips, local humidity gradient changes need to be captured.

[0081] A thin-film NTC humidity sensor II-5 is placed in the palm area to monitor the overall humidity of the palm.

[0082] An environmental sensor II-6 is set at the auxiliary monitoring point on the wrist to distinguish between ambient humidity penetration and sweat evaporation inside the glove, with a spacing of about 15 cm.

[0083] To prevent sweat accumulation from affecting the operation of the internal sensors of the gloves, an airflow guidance design is adopted, and guide grooves II-8 are set at the seams of the glove body to ensure air circulation around the humidity sensor and reduce measurement errors caused by dead volume.

[0084] The layered structure of each temperature / humidity sensor mentioned above is as follows Figure 7 The flexible precision temperature and humidity sensor II-4 includes, arranged from top to bottom, a circular flexible encapsulation layer II-4-1, a circular flexible thermal-sensitive material upper electrode II-4-2, a circular flexible thermal-sensitive material II-4-3, a circular flexible thermal-sensitive material lower electrode II-4-4, a circular flexible humidity-sensitive material upper electrode II-4-5, a circular flexible humidity-sensitive material II-4-6, a circular flexible humidity-sensitive material lower electrode II-4-7, and a flexible base layer II-4-8.

[0085] The thin film NTC temperature and humidity sensor Ⅱ-5 includes a rectangular packaging layer Ⅱ-5-1, a rectangular upper electrode Ⅱ-5-2, a rectangular temperature-sensitive functional layer Ⅱ-5-3, a rectangular humidity-sensitive functional layer Ⅱ-5-4, a rectangular lower electrode Ⅱ-5-5 and a rectangular substrate Ⅱ-5-6, which are arranged in sequence from upper to lower layers.

[0086] The environmental temperature and humidity sensor II-6 includes a strip-shaped packaging layer II-6-1, a strip-shaped thermal conductive layer II-6-2, a strip-shaped thermistor II-6-3, a strip-shaped humidity-sensitive material upper electrode II-6-4, a strip-shaped humidity-sensitive material II-6-5, a strip-shaped humidity-sensitive material lower electrode II-6-6 and a strip-shaped base layer II-6-7, which are arranged in sequence from the upper layer to the lower layer.

[0087] The microclimate within the glove is influenced by finger movement, respiratory exchange, and the external environment. A three-level temperature detection gradient layout, spanning fingertips, palms, and wrists, creates a monitoring network with decreasing spatial resolution. High-density fingertip deployment captures minute temperature and humidity fluctuations caused by localized sweat evaporation, while palm sensors monitor overall thermal equilibrium. Wrist sensors filter out environmental interference, forming a three-dimensional sensing system that integrates "point-surface-volume."

[0088] Secondly, multi-physics field coupling analysis reveals a strong correlation between temperature and humidity (for example, sweat evaporation absorbs heat, causing localized cooling). Monitoring either parameter in isolation can easily lead to misjudgments. The combined sensor uses data fusion algorithms (such as gray correlation analysis) to establish a temperature-humidity coupling model. This model can distinguish between physiological sweating (stable temperature / sudden humidity increase) and ambient moisture penetration (synchronous changes in temperature and humidity), significantly reducing false alarm rates.

[0089] The main control unit of this embodiment is used to issue a third-level alarm message when the humidity is greater than a first set humidity value, or the temperature is greater than a first set temperature threshold, and the duration is greater than a set time; when the temperature is greater than a second set temperature threshold, it is used to issue a second-level alarm message.

[0090] Among them, the first set humidity value is 85%.

[0091] The first set temperature threshold is 34°C; the second set temperature threshold is 45°C.

[0092] The third-level alarm information includes sound and light alarm information, and the control air pump stops working.

[0093] Among them, the main control unit is also used to determine that when the humidity of a certain part is greater than the first set humidity value, based on the temperature and humidity of the part, the cause of the increased humidity is the physiological sweating, and only when the cause of the increased humidity is physiological sweating, a third-level alarm information is issued; specifically, when the humidity of a certain part is greater than the set value and the temperature and humidity change synchronously, it is determined that the cause of the increased humidity is the infiltration of environmental moisture; when the humidity of a certain part is greater than the set value, but the temperature and humidity change asynchronously, it is determined that the cause of the increased humidity is physiological sweating.

[0094] The reason is that when the human body actively perspires to regulate its temperature, the skin's surface temperature typically remains stable or slightly decreases. Sweat secretion causes a rapid increase in local humidity, but the temperature change is relatively small. This results in a mismatch between the changing trends of temperature and humidity, such as a sudden increase in humidity while the temperature remains stable.

[0095] External high-humidity environments (such as rainy days and bathrooms) are often accompanied by temperature fluctuations (for example, the temperature may drop when the air humidity increases). Humidity gradually penetrates the sensor surface, causing the humidity to slowly rise. At the same time, the temperature and humidity change synchronously. In this case, the changing trends of temperature and humidity are highly correlated, such as the temperature and humidity rising or falling synchronously.

[0096] This embodiment improves health monitoring accuracy and avoids false alarms by determining the cause of elevated humidity. By distinguishing between different types of sweating, it filters out environmental interference and improves alarm accuracy. The main control unit only issues an alarm when it determines that elevated humidity is caused by physiological sweating. In non-physiological sweating scenarios, the sensor sampling frequency is reduced or redundant functions are disabled to extend battery life. Users will not receive frequent alerts due to false alarms in humid environments.

[0097] The main control unit processes humidity and temperature data to determine whether to issue an alarm. The process includes: (1) Preprocessing of temperature and humidity, including filtering and denoising, using sliding average filtering or Savitzky-Golay filtering to eliminate sensor noise, and correcting the temperature drift error of the humidity sensor (such as the nonlinear effect of humidity changing with temperature) through temperature compensation algorithm.

[0098] (2) Calculate the temperature change rate, humidity change rate and temperature and humidity correlation of the pre-processed temperature and humidity data ; The rate of temperature change is equal to , the humidity change rate is equal to ,The correlation between temperature and humidity is characterized by grey ,relationship, where T is temperature, t is time, and RH is humidity.

[0099] Grey relational analysis (GRA) Calculate the correlation between temperature and humidity , distinguish whether the temperature and humidity change trends are synchronized:

[0100] It is the correlation degree, which is used to measure the synchronization of temperature and humidity change trends, with a value range of 0~1. , indicating that the temperature and humidity changes are decoupled, which is physiological sweating. is the synchronous change of temperature and humidity, which is the ambient moisture penetration or high temperature; n is the total number of data points, that is, the total number of temperature and humidity data pairs involved in the calculation; i is the index of the traversal temperature data sequence, i=1,2,...,n; j is the index of the traversal humidity data sequence, j=1,2,...,n. is the temperature data sequence, i.e. the original temperature value at the i-th moment, is the humidity data sequence, i.e. the original humidity value at the jth moment, It is the resolution coefficient, usually 0.5, which is used to adjust the sensitivity of the correlation, avoid the denominator being zero, and enhance the formula's ability to resolve small differences.

[0101] (3) Determine whether to issue an alarm based on temperature, humidity, temperature change rate, humidity change rate, and the correlation between temperature and humidity.

[0102] Specifically, the judgment logic for whether to issue an alarm message is: when the humidity is greater than the first set humidity value, the temperature is less than the first set temperature threshold, and the correlation between temperature and humidity is less than the set first correlation threshold, it is determined that the cause of the increase in humidity is physiological sweating, and a third-level alarm message is issued; when the temperature is greater than the first set temperature threshold, the humidity is greater than the second set humidity value, and the correlation between temperature and humidity is greater than the set second correlation threshold, it is determined that the cause of the increase in humidity is environmental moisture penetration, and no alarm message is issued; when the temperature is greater than the set second set temperature threshold, it indicates that there is a risk of burns, and a second-level alarm message is issued.

[0103] Preferably, the temperature, humidity, temperature change rate, humidity change rate and the correlation between temperature and humidity are analyzed by a decision tree classification model to determine whether to issue an alarm message.

[0104] Among them, the input of the decision tree classification model is temperature, humidity, temperature change rate, humidity change rate and the correlation between temperature and humidity; the judgment logic of whether to issue an alarm information is used to analyze the input data, determine whether to issue an alarm information, and output it.

[0105] Alarms are triggered only when parameters exceed the limit continuously, avoiding false alarms due to instantaneous fluctuations.

[0106] like Figure 1 As shown, the pneumatic system includes the glove body III-3, the air pump IV-5 and the pipeline III-1. The air pump IV-5 is connected to the pipeline III-1, and the pipeline III-1 is fixed to the glove body III-3. It can realize independent driving of a single finger, and the five fingers are interconnected. The movement of different fingers will drive the movement of the other fingers. Among them, the glove body III-3 is made of breathable and soft materials (such as PDMS, silk fiber). The pipeline III-1 is composed of a high-strength elastomer (such as polyurethane TPU) and a spiral fiber braided layer. By filling the pipeline with compressed air, the radial expansion and axial contraction deformation of the elastomer are caused, thereby controlling the movement of the glove body; a connecting knot III-2 is set at the connection of the pipeline III-1, and the smooth flow of the pipeline is ensured by the connecting knot III-2. Its key advantages are: Flexible drive: no rigid connecting rod structure, avoiding the risk of secondary damage.

[0107] Passive rebound: automatically returns to its original shape after pressure relief, with high energy recovery rate.

[0108] Lightweight: More than 80% lighter than traditional cylinders, and lighter per unit length.

[0109] Among them, pipeline III-1 includes five finger pipelines; air pump IV-5 includes five finger air pumps and a main air pump; the five finger pipelines are connected to the five finger air pumps one by one; the five finger pipelines are also connected to the main air pump; the five finger pipelines are inflated at the same time by the main air pump to control the synchronous extension or opening of the five finger pipelines; the corresponding finger pipelines are inflated by a separate finger air pump to control the extension or opening of the separate finger pipelines.

[0110] In this embodiment, the air pump, energy storage unit and main control unit are all installed on the wristband IV-4, and the wristband IV-4 is also provided with a display screen IV-1. Figure 8 shown.

[0111] Display screen IV-1, main control unit IV-2, battery IV-3, wristband IV-4, and air pump IV-5 are integrated into integrated wristband IV, which can be worn on the wrist. The main control unit is also connected to a mobile client for data exchange. The mobile client can be a mobile phone, computer, etc. Figure 9 This is a mobile app interface. The upper left corner displays the ambient temperature, and the upper right corner displays the glove battery level. The first screen is the homepage, displaying static data such as battery level and whether training has taken place that day. The second screen is the training mode, with four available modes: rehabilitation and precision training. The third screen is the data monitoring screen, displaying data uploaded by various sensors, such as the bending condition of each finger and the temperature and humidity of the glove. The fourth screen is the training record, which stores historical training records, including training date, training mode, and training time.

[0112] Figure 10 This is the overall workflow diagram of the flexible hand rehabilitation training system based on multimodal sensing and self-powering disclosed in this embodiment. Thermoelectric generation layer I-3 powers the battery, which in turn provides power to the sensors, air pumps, control unit, and other components. The air pump drives the corresponding pipelines to control the bending of each finger. The bending status is fed back to the control unit by the bending sensor and the IMU sensor. Simultaneously, the temperature and humidity sensors collect the temperature and humidity of the glove and feed it back to the control unit. Because the system uses hand heat for self-powering, it ensures timely and stable power consumption while reducing its size. The sensors are weighted according to the data they collect: the pressure sensor is weighted 45%, the IMU sensor is weighted 30%, the temperature sensor is weighted 15%, and the humidity sensor is weighted 10%. Finally, all data is uploaded to the mobile phone for easy viewing.

[0113] By assigning weights to each sensor, data importance can be graded. Among them, the data obtained by the pressure sensor directly reflects the mechanical state of finger flexion and extension, such as muscle strength, movement accuracy, etc., and is the core data for evaluating the effectiveness of rehabilitation training.

[0114] The IMU sensor acquires data to provide posture calibration and motion continuity monitoring, such as finger rotation angle and acceleration, and assists in the spatiotemporal alignment of pressure data.

[0115] The data obtained by temperature / humidity sensors is mainly used for safety warnings such as burn risks and environmental interference. It has a lower weight but is indispensable.

[0116] When transmitting data, priority is given to ensuring the transmission of high-weight data. Pressure sensor data has high real-time requirements and needs to be transmitted to the mobile phone first; temperature / humidity data can reduce the sampling frequency or be cached locally.

[0117] The operating mode of the sensor can also be dynamically assigned according to the weight (for example, the IMU only activates high-frequency sampling when an action occurs).

[0118] Example 2 In this embodiment, a flexible palm rehabilitation training method based on multimodal sensing and self-power supply is disclosed, including: The palm heat power supply system is based on the thermoelectric effect, converting skin temperature into electrical energy to power the energy storage unit; Power the air pump and main control unit through the energy storage unit; The main control unit controls the air pump to inflate the pipes, thereby controlling the movement of the glove body.

[0119] Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. Flexible palm rehabilitation training system based on multimodal sensing and self-powered, characterized by: Including pneumatic system, palm heat power supply system, energy storage unit and main control unit; The pneumatic system includes a glove body, an air pump, and a pipeline. The air pump is connected to the pipeline, which is fixed to the glove body. The air pump and the main control unit are both connected to the energy storage unit. The air pump and the main control unit are in communication connection, and the energy storage unit is connected to the palm heat power supply system. The palm thermal power supply system can convert skin temperature into electrical energy based on the thermoelectric effect to power the energy storage unit.

2. The flexible palm rehabilitation training system based on multimodal sensing and self-power supply according to claim 1, characterized in that: It also includes a multimodal sensing system, which is connected to the palm thermal power supply system and the main control unit; The multimodal sensing system is used to obtain the temperature and humidity of each area of ​​the glove body; The main control unit is used to issue an alarm message when the temperature or humidity exceeds the set threshold.

3. The flexible palm rehabilitation training system based on multimodal sensing and self-power supply according to claim 2, characterized in that: The main control unit is used to issue a third-level alarm message when the humidity is greater than the first set humidity value, or the temperature is greater than the first set temperature threshold, and the duration is greater than the set time; when the temperature is greater than the second set temperature threshold, it is used to issue a second-level alarm message.

4. The flexible palm rehabilitation training system based on multimodal sensing and self-power supply according to claim 3, characterized in that: The main control unit is also used to determine that when the humidity of a certain part is greater than the first set humidity value, based on the temperature and humidity of the part, the cause is the increased humidity. Only when the cause of the increased humidity is physiological sweating will a third-level alarm message be issued.

5. The flexible palm rehabilitation training system based on multimodal sensing and self-power supply according to claim 1, characterized in that: It also includes a wristband, on which the air pump, energy storage unit and main control unit are all installed. A display screen is also provided on the wristband, and the display screen is connected to the main control unit.

6. The flexible palm rehabilitation training system based on multimodal sensing and self-power supply according to claim 1, characterized in that: The main control unit is connected to the mobile client.

7. The flexible palm rehabilitation training system based on multimodal sensing and self-power supply according to claim 1, characterized in that: The palm thermal power supply system includes a contact thermal conductive layer, a thermoelectric generator array layer, a heat dissipation layer and a packaging layer stacked in sequence; wherein the contact thermal conductive layer is used to contact the skin.

8. The flexible palm rehabilitation training system based on multimodal sensing and self-power supply according to claim 1, characterized in that: The thermoelectric power generation layer includes a flexible substrate and a TEG unit array arranged on the flexible substrate; the TEG unit array in the thermoelectric power generation layer includes three connection layers; wherein the first connection layer includes two connected TEG units, and a switch S is also provided on the connection circuit of the two TEG units P1A and S S1A , by controlling the switch S P1A and S S1A The second connection layer includes two connected first connection layers, and a switch S is set on the two connected first connection layers. P2A and S S2A , by controlling the switch S P2A and S S2A The two first connection layers are connected in series or in parallel by opening and closing; The third connection layer includes three connected second connection layers, on which switches S are set. P3A 、S S3A and S S3B , by controlling the switch S P3A 、S S3A and S S3B The three second connection layers are connected in series or in parallel by switching them on and off.

9. The flexible palm rehabilitation training system based on multimodal sensing and self-power supply according to claim 1, characterized in that: The pipeline includes five finger pipelines; the air pump includes five finger air pumps and a main air pump; the five finger pipelines are connected to the five finger air pumps in a one-to-one correspondence; and the five finger pipelines are also connected to the main air pump.

10. A flexible palm rehabilitation training method based on multimodal sensing and self-powered power supply, characterized in that: include: The palm heat power supply system is based on the thermoelectric effect, converting skin temperature into electrical energy to power the energy storage unit; Power the air pump and main control unit through the energy storage unit; The main control unit controls the air pump to inflate the pipes, thereby controlling the movement of the glove body.

Images